Brain 'imaging'with polytrodes: precise 3D neuron localization and classification with silicon electrode arrays. Single tungsten electrodes have been the principal technology for in vivo electrophysiological studies for decades, yet they have significant limitations. In addition to the low yield in obtaining stable single-unit recordings, this technique provides limited information about the location and type of the recorded cell, and no information about the connectivity between neurons. A promising recent technological advance is silicon-substrate multielectrode arrays ('Michigan probes', or 'polytrodes'). Compared to conventional single- or multi-wire electrodes, polytrodes provide better single-unit isolation and enable simultaneous recording from dozens of neurons. The polytrodes central to our proposal have 54 closely spaced recording sites and can record and isolate up to 150 neurons simultaneously. We recently developed algorithms that capitalize on the closely spaced site geometry to localize neurons in 3D cortical space using a monopole-dipole field model of extracellular spike potentials. In the proposed project we will combine polytrode recordings with 2-photon Ca2+ imaging to validate and refine these algorithms. The goal is to establish a technique for simultaneous recording of hundreds of neurons within a neocortical column in vivo, one that provides: (a) precise 3D localization and cell type classification of the ensemble of neurons, (b) characterization of neuronal response properties (e.g. receptive fields) in the context of the cortical layers, and (c) measurement of synaptic connectivity. These experiments will provide a definitive validation of the model so that in vivo 3D functional images of cortical columns can be made independently of 2-photon imaging. Compared to the currently available techniques for imaging population activity, polytrodes provide higher temporal resolution, allow access to all cortical layers, and are more readily applicable to awake behaving animals. With conventional electrophysiology, while neuronal response properties such as receptive fields are often studied in vivo, the location, cell type, and synaptic connectivity of neurons are typically only characterized in vitro. The technique we propose provides complementary information about the functional and anatomical properties of a large number of neurons in vivo, bridging the divide between electrophysiology and cellular histology. It promises to be a powerful new tool for researchers studying the neuronal circuitry of normal brain function and neurological disease. PUBLIC HEALTH RELEVANCE: Our understanding of how networks of neurons in the brain interact has been limited by conventional electrodes that can only record one neuron at a time, without identifying the type or anatomical location of the neuron. By combining multi-electrode arrays with computer modeling, we propose a novel technique for monitoring the activity of hundreds of neurons that not only identifies the precise location of each neuron, but also distinguishes between different cell types. In addition to furthering the development of brain-machine interfaces for neural prosthetics, this technique will provide a powerful new tool for studying neurological diseases such as epilepsy, where simultaneous recording of many neurons and identification of cell type is key to deciphering the neural basis of this debilitating condition.